Steven Chu looks down the road and sees an America that is free from foreign oil, powered by home-grown genetically engineered fuel that burns cleanly and is as available as the weeds that grow in your garden.
And where does this Nobel laureate get his inspiration? From termites. Or more specifically: the guts of termites.
Chu, who won the Nobel for physics in 1997, isn't claiming that termites can save the world, as one headline recently screamed. But he does believe that the natural processes that allow termites to turn the hard fabric of plant material -- cellulose -- into an ethanol-like fuel hold secrets that could lead to cheap, clean-burning and virtually limitless fuel.
But first, he has to figure out how the termite works its magic.
"I haven't a clue," he says candidly about exactly what goes on in the guts of a termite. "But I'll find out."
Chu isn't thinking about building a better mouse trap. He's thinking about building a better termite. Or more precisely, he wants to create a whole new class of microorganisms, tiny microbes that are genetically engineered to produce far more fuel, or ethanol, than they need to survive. They would be tailor-made versions of the microbes that convert cellulose to fuel for termites.
It sounds far out. But this isn't a guy given to daydreaming. He has challenged the prestigious lab he now heads, Lawrence Berkeley Laboratory, to attack the problem on a multi-disciplinary basis.
"This is Steve's crusade right now," says a colleague. "He feels passionate about it."
Ethanol is a hot-button issue today, and hotly debated, because it can be used as a substitute for gasoline. But there's one serious problem. According to various studies, it takes more energy to extract ethanol from the most common source, corn, than the amount of energy that ends up in the ethanol.
It's a net energy loser, and while the fuel may be relatively clean, the processes required to produce it aren't.
It's "clearly a bad bargain," says one report out of the Berkeley lab.
That's precisely what Chu wants to change.
"We're beginning to sequence the microorganisms in the gut of a termite," he says. Once the genes are identified, the scientists can develop a "chemical understanding of what's really happening in breaking down the cellulose."
Cellulose forms the stiff walls of plant cells, so it is present in all plant material, even the mountains of trash that end up in landfills. It can be broken down, or fermented, by microbes to produce various products, like hydrogen or methanol.
But unfortunately, no microbes produce enough of it to be useful to humans. And that's exactly what Chu wants to change.
"If you look at how nature degrades biomaterial, like cellulose, you find that there are microorganisms that make things like ethanol and methane and so on," he says. But these microbes only make enough fuel to satisfy their own needs.
"We can actually begin to take these organisms that do this in nature and get them to make more of the stuff they're already making. This goes beyond normal evolution. So the question is can we stimulate evolution along a particular path?" Chu says.
He thinks the answer is yes.
"We want to make conditions such that the microorganism that is a little more efficient [at degrading cellulose] will be encouraged to survive while the less-efficient ones are not encouraged to survive. Things like this are beginning to happen.
"So now imagine an organism that has to survive by making hydrogen or methane. And you alter the conditions so you are asking it to make more and more of the stuff. You can do this in gradual steps. I want a certain property, and in order to survive it has to make this property. Gobs of methane. Not just a little bit."
That might be done by selective breeding, like that used to produce faster racehorses or better corn, or by inserting a special code into the DNA of the microbe.
"You can actually insert into the genome of the microorganism a set of genes that will allow it to do the chemistry you want. It's called synthetic biology," Chu says. "You can insert the entire chemical cycle that you want into this organism."
That would require maintaining a careful balance, because the microbes probably won't be all too happy about this.
"They are not really feeling so well because they are making something they don't really want to make," Chu says. "They don't really want to make it, but they're making it. If they make too much of it they're going to die because they're not making enough of the other stuff that they need to live. So it's a threshold thing. You want them to be healthy and multiply but you also want them to do what we want them to do."
Chu knows not everybody out there is going to be comfortable with this approach to energy salvation.
"This has sort of a Frankenstein air to it," he admits. "In principal you could have a situation where microorganisms go berserk , but there could be ways around that."
The microbes could be programmed, for instance, to die if they ever escape.
"You can imagine doing something where they can only survive in a very artificial environment," he says, but he knows some critics won't be content with that.
"There are always people who are going to be upset about this because quite frankly it's really new, they don't understand it, and it's scary."
For the record, Chu is not out there all by himself on this. Other institutions, including the University of Colorado, the U.S. Department of Energy's National Renewable Energy Laboratory, and others, are making headway on the same issues.
It will take awhile, of course, but Chu's timetable is more optimistic than some.
"This is a 10- to 20-year project," he says.
If he's on the right track, someday the world might be free of energy shortages, and wars over oil, and soaring prices at the gas pump. And we might have a cleaner environment.
That's enough to give a termite something to chew on.
Lee Dye's column appears weekly on ABCNEWS.com. A former science writer for the Los Angeles Times, he now lives in Juneau, Alaska.